MLOps with AWS - Bootcamp - Zero to Hero Series

Learn how mlops standardizes the machine learning lifecycle and streamlines production deployment. Collaborate across data engineers, data scientists, ml engineers, and stakeholders to ensure governance, monitoring, and data quality.

Explore the end-to-end MLOps pipeline, from data preprocessing to deployment and production monitoring, illustrated with a live use case.

Learn mlops fundamentals, including development, staging, and production environments, assets like code files and model registry, and an AWS-powered CI/CD pipeline using code commit, code build, and SageMaker.

Explore the end-to-end mlops workflow—from data sets and feature tables to feature store, model training, deployment, monitoring, and retraining, powered by ci/cd pipelines and production guardrails.

Address the data-driven, iterative nature of machine learning with MLOps, which requires specialized infrastructure and governance, and explain why DevOps alone cannot effectively manage training, deployment, and data-focused workflows.

Discover what AWS is, its pay-as-you-go cloud services, and how it enables scalable, on-demand compute, storage, and databases. Learn IaaS, PaaS, SaaS, and deployment models.

Explore the AWS machine learning stack for MLOps, from data prep and feature store to SageMaker capabilities, training, deployment, and governance. See how pay-as-you-go services accelerate end-to-end ML.

Understand software development life cycle, a structured process that delivers high quality software efficiently. Explore SDLC phases from planning to monitoring, with deliverables, estimates, and risk reduction for machine learning.

Explore the main SDLC models—waterfall, agile, V-model, lean, iterative, and spiral—through their structure, strengths, and when change, feedback, and risk drive development.

Compare waterfall, agile, and DevOps models and explore how AWS tools enable DevOps in the mlops lifecycle, highlighting iteration, automation, and cross-team collaboration.

Discover how DevOps combines culture, practices, and AWS tools to deliver software fast through continuous integration and delivery, infrastructure as code, microservices, and cross-functional collaboration with automated monitoring.

Learn AWS basics for beginners in the MLOps with AWS bootcamp, including creating an account, IAM users with MFA, IAM policy generator, and using the CLI, S3, and EC2.

Create your AWS free tier account, review free tier services like EC2 and Sagemaker, understand potential charges, and complete account setup with verification and billing details.

Learn how to set up multi-factor authentication on the AWS root account, configure IAM security settings, and enable MFA with authenticator apps like Google Authenticator.

Create an IAM user with console access, attach administrator privileges, and use an account alias for sign-in; log in as the IAM user and set up MFA for security.

Install AWS CLI v2, verify with aws --version on Mac OS or Windows, then configure access key id and secret access key and set us east one with json output.

Learn to manage access with IAM by creating users, attaching policies, and granting explicit permissions. Explore service roles and execution roles used for service calls in AWS.

Explore how to generate and attach IAM policies in AWS, using policy generator to create deny and allow statements for S3, and attach inline policies to a user.

Learn how to delete an IAM user and associated service roles in AWS by selecting the user or role, confirming deletion, and removing them from your account.

Learn how Amazon S3 uses buckets and objects with encryption and internet access. Compare storage classes such as S3 Standard, S3 Standard IA, S3 Intelligent-tiering, and S3 Glacier.

Create an S3 bucket from the console with a globally unique name, configure region and permissions, enable or disable versioning, manage public access, and upload an object.

Create a globally unique S3 bucket via the CLI using a random ID, upload a file, view storage class, and delete the bucket.

Explore S3 versioning by creating a bucket, uploading a file twice, enabling versioning, and viewing, restoring, or permanently deleting previous object versions via the AWS CLI.

Explore EC2, the elastic compute cloud providing secure, scalable on-demand compute capacity. Use GPU powered instances for ML, auto scaling, and EC2 fleet on AWS global infrastructure.

Launch an EC2 instance using the Amazon Linux operating system, select Linux 2 AMI and t2.micro, create a key pair, configure security groups for SSH, then connect via SSH.

Go back to the EC2 instances page, select the running server, and terminate the instance to stop billing. As a cleanup, delete any created S3 buckets.

Learn the basics of the Linux operating system and command line interface for managing Linux-based servers in DevOps and data science, using an EC2 instance with Amazon Linux 2 AMI.

Explore the basics of Linux, its core components—bootloader, kernel, init, daemons, the X server, and desktop environments—plus Bash and package managers across popular distros.

Launch and configure an EC2 instance in the AWS console, selecting Amazon Linux 2 or Ubuntu, t2.micro free tier, a key pair, security group, and 10 GB storage.

Master Linux basic commands for data science and DevOps, covering directory operations, file creation and editing, and permissions.

Understand the ci cd pipeline, including code, build, test, provisioning, deployment, and monitoring, and how continuous integration and delivery automate deployment.

Learn that AWS CodeCommit is no longer available for new users, and transition to GitHub for storing git repositories, with a dedicated hands-on for MLOps.

Explore how aws code commit and dvcs enable secure, distributed version control with private git repositories, encrypted in transit and at rest, as a scalable managed service.

Install and configure git across macOS, Linux, and Windows, verify version, set user name and email, enable color output, and adjust auto crlf to prevent cross-platform line endings.

Set up a git workflow with AWS code commit by creating a remote repository and pushing changes from the local workspace to the remote repository.

Learn the git workflow from a working directory to pushing commits to AWS code commit, covering staging, local repositories, and master and feature branches.

Learn how git add moves files from the working directory to the staging area. Use patterns to add all files, specific lists, or .txt extensions.

Learn how files move from the working directory to the staging area and local repository, and use git diff and git status to visualize differences across stages and commits.

Learn to unstage changes across the working, staging, and commit areas using git reset head and soft resets, and review differences with git diff.

Learn how to move files between staging, commit, and working areas using git reset and git revert, and master techniques like git commit amend and soft or hard resets.

Connect your local repo to aws code commit by adding a remote named origin and pushing changes with git push -u origin master.

Clone and branch git repositories by copying from remote to local, configuring remote links with git remote -v, and pushing or pulling changes, while managing remotes in a website-building project.

Perform hands-on git branching from master to a dev branch, switch branches in VS Code, commit and push changes to AWS CodeCommit, and explore branching and merging concepts.

Learn to manage feature work with git branches: develop on a dev branch, commit and push to code commit, then merge into master with a fast-forward merge.

Learn to identify and resolve git merge conflicts when integrating parallel feature branches into master, using manual conflict resolution in index.html, then commit and push changes.

Learn to differentiate git rebase from git merge, achieve a linear history with rebase, and handle fast-forward merges and conflicts in feature and master branches.

Learn how to use git stash save to save uncommitted work on a feature branch, switch to master, and manage stashes with list, apply, pop, drop, and clear.

Learn practical git stash techniques to manage unfinished changes across branches, including stash save, list, apply, pop, drop, and clear, with real-world examples on master and feature branches.

Secure AWS code commit by applying explicit deny policies with IAM, restricting master and production branches to senior developers and allowing juniors to work on their own branches.

Set up AWS code commit security with an explicit deny policy on master for junior developer group using an inline policy. Verify dev pushes succeed and master pushes are blocked.

Learn how AWS code commit integrates with triggers, notifications, and CloudWatch events (EventBridge), and automate builds by routing changes to SNS topics, CodeBuild projects, and Lambda.

Explore how AWS CodeCommit provides a secure, scalable repository for hosting private git repositories, enabling team collaboration via branches in a managed CI pipeline.

Learn how Git, a distributed version control system, tracks changes for ML projects. Use branches, CI/CD, and infrastructure as code to enable collaboration and reproducibility in MLOps.

Differentiate local and remote repositories and learn to work with git-tracked code on the cloud using hosting services such as GitHub, GitLab, Bitbucket, and CodeCommit.

Check for an existing global git configuration, create it if missing, and set user name, email, and init.defaultBranch as main; verify with git config --global --list.

Initialize a local repository by running git init inside a project folder, turning it into a git-managed directory with a hidden dot git folder that stores changes.

Learn how a working directory becomes a repository with git init, then use the staging area and commits to create versioned snapshots of your ml project tracked in commit history.

Install git on your local system and verify with git version. Use git bash on Windows or homebrew on macOS, and set up VS Code to explore the repository.

Learn to manage local files in a git workflow by creating, staging, and committing a Python file in a local repository. Track changes with git status and git log.

learn how git branches are movable pointers to commits, create a dev branch from main, and see main and dev point to the v2 commit as head shows active branch.

Switch git branches using git switch or git checkout, moving from main to development and tracking the head, status, and latest commit.

Master git merging by understanding fast-forward merges and three-way merges, and see how merge commits reflect combining branches like main and development.

Understand how to check out commits in git, switch to a specific commit, and manage head state and uncommitted changes for debugging and branching.

Explore git hosting services with a focus on GitHub and set up credentials. Generate a personal access token, configure permissions and expiration, and prepare to create repositories and push code.

Learn to work with remote repositories to enable collaboration, push and pull changes, and maintain version history, backups, and access control within a continuous integration and deployment workflow.

Learn how to clone a git repository, create and push new branches, and delete branches locally and remotely to manage code across remotes.

Learn how three way merge in git combines changes from two branches using a common base commit, branch a and branch b tips, and a merge commit, including conflict resolution.

Explore git as a distributed version control system for source code management. Track changes, enable collaboration, and master local and remote repositories, branching, and three way merge.

explore yaml crash course and learn how yaml, a data serialization language, is used to write human-readable configuration files with indentation rules, keys, maps, lists, and multi-line strings.

Learn how CodeBuild, a fully managed CI service, compiles code, runs tests, and packages deploy-ready software, while generating a Docker file for Amazon ECR and scaling without provisioning.

Learn to create your first AWS CodeBuild project by integrating CodeCommit, configuring the source and environment, and using a buildspec.yaml to drive build phases with CloudWatch logs.

Connect a GitHub repository to AWS CodeBuild, create a CodeBuild project, push code, configure credentials, and run a build with a buildspec.yml.

Follow along as the instructor covers two code build variants: one uses environment variables with a parameter store, and another hands-on session pushes artifacts to an S3 bucket.

Drive into buildspec.yml deep dive, learning how to define build commands in YAML, manage install, pre-build, build, and post-build phases, and use env vars, artifacts, and parameter store with codebuild.

Explore AWS CodeBuild hands-on by modifying index.html to fail a condition, push to CodeCommit, start a new build, and analyze a failure, plus timeout and environment configuration.

Learn to define and access environment variables in AWS CodeBuild using buildspec.yml, plain text, parameter store, and secrets manager; override builds and securely retrieve dynamic values with proper IAM permissions.

Learn how to save AWS CodeBuild artifacts to an S3 bucket by configuring artifacts in CodeBuild and buildspec.yml, including folder build app and proper line endings.

Automate builds by linking AWS CodeCommit to CodeBuild via EventBridge. Trigger on commit, pull latest code, then push changes and monitor build logs for end-to-end automation.

Perform a quick cleanup by deleting the code commit repository, build project and builds, removing the EventBridge rule, and deleting any S3 buckets created during this learning journey.

Learn how AWS CodeDeploy automates application deployments across EC2, Fargate, Lambda and on-prem environments, enabling blue-green updates, rollback, and zero downtime in CI/CD pipelines.

Create EC2 instances with IAM roles and the CodeDeploy agent, push a versioned code revision to S3, and deploy it to a deployment group using AWS CodeDeploy.

Create and deploy a code deploy workflow on aws by provisioning ec2 instances with iam roles, installing the code deploy agent, and deploying from s3 using a deployment group.

Explore the appspec.yml deep dive for AWS CodeDeploy, detailing lifecycle hooks, YAML versus JSON formats, and EC2 deployment with CodeDeploy agent, S3 bucket integration, and deployment configuration options.

Explore how CodeDeploy automates deployment to compute instances and how CodePipeline orchestrates CI/CD, builds, tests, and artifact generation from commits to S3.